Modelling and forecasting monthly palm oil prices in Nigeria: (1960 – 2024)

Abstract

This study investigates the dynamics of palm oil prices in Nigeria from January 1960 to May 2024, with a focus on identifying patterns and developing an effective forecasting model. Descriptive statistics revealed significant variability in prices, with an average price of $532.58 and a high coefficient of variation at 52.27%, indicating substantial fluctuations over time. The price distribution exhibited positive skewness and high kurtosis, suggesting the presence of extreme values. To model and forecast these prices, a Seasonal Autoregressive Integrated Moving Average (SARIMA) model, specifically SARIMA (2,0,1) x (2,0,2)₁₂, was selected based on its ability to capture both the autoregressive and seasonal components of the data. The model's performance was evaluated through several metrics, including RMSE, MAE, and MAPE, all of which indicated a high level of accuracy in forecasting future prices. However, some degree of unexplained variability remained, pointing to external factors not accounted for by the model. The findings underscore the complexity of the palm oil market in Nigeria and highlight the importance of incorporating both past values and seasonal effects in price forecasting. The study's results have significant implications for policymakers and industry stakeholders, suggesting the need for targeted interventions to manage price volatility and improve market stability. Recommendations for future research include the incorporation of additional external variables and the exploration of alternative modeling techniques to enhance forecast accuracy.